Printer Friendly

The FHA facts.


A detailed look at FHA insurance and claims data reveals much about the health of recent years' originations.

Statistics recently made available by the Department of Housing and Urban Development on the performance of its FHA-insured portfolio reveal some encouraging signs for the mortgage lending industry. This article portrays some of the prominent results from the HUD study of FHA insurance and claims data, which covers the performance of Section 203(b) 30-year term mortgages originated from 1970 to 1991 as of June 30, 1991. From this data, several salient points can be observed: * More than two-thirds of FHA's portfolio

was originated in 1986 or later.

Only 19 percent of the loans originated

in the 1970s remain on FHA

books. * Loans originated in the 1986-to-1988

period are performing 100 percent

better than the 1980-to-1985 books

of business over similar periods of

time and have claim rates comparable

to the 1970s' books of business. * To no one's surprise, FHA claims

have been concentrated in the "oil-patch"

states during the 1980s.

Forty-eight percent of the claims

HUD has paid on loans originated in

fiscal years 1985 to 1986 came from

Texas, Oklahoma, Louisiana, Colorado

and Alaska. Fifty

percent of all FHA

loans originated in

the state of Alaska in

the first half of the

1980s have already

resulted in claims for


Now we can observe the various ways this data breaks down and what impact it holds for the mortgage lending industry.

Performance of

recent business

During the 1980s, FHA experienced a tremendous increase in activity. FHA insured 4.2 million mortgages from 1986 to 1990 - more than it did in the previous 13 years from 1973 to 1985. It is not surprising, therefore, that of FHA's insurance written since 1970 that is still in force, 71 percent representing 3,278,359 insured loans, was originated in 1986 or later. Table 1 provides specific data broken down into four-year origination periods on the seasoning of FHA's portfolio as of June 1991.

Table : TABLE 1 Seasoning of FHA's Portfolio (Insurance-In-Force as of June 1991)
Years of Mortgages Percent
Origination Outstanding Of Total
1971-75 315,593 7
1976-80 602,930 13
1981-85 410,316 9
1986-90 3,278,359 71

FHA's future solvency will be dependent on the performance of FHA origination activity since 1986. While it is desirable to have a more-seasoned portfolio from a risk management perspective, the concern about the concentration of recent originations is abated by the fact the FHA has tightened its underwriting policies since 1986 with over 30 different measures (e.g., buydown restrictions, elimination of risky programs and stricter compensating factors for borrowers above credit ratio guidelines) to reduce credit risk. In addition, many of the loans originated in the last half of 1986 and much of 1987 are refinances of loans originated in earlier years.

Claim performance improving

Chart 1 compares the cumulative claim rates for FHA loans originated from 1970 to 1988 for the first three policy years. The chart documents the rise in claim rates for loans originated in the early 1980s. Claim rates for loans originated in 1986 to 1988 are lower and, in fact, are comparable to the levels of claim rates in the 1970s. Accordingly, while FHA's portfolio is unseasoned, the improved performance of recent originations is very encouraging.

In trying to pinpoint the causes of this performance over the last 20 years, Chart 2 shows a direct correlation between HUD's cumulative claim rates by year of insurance endorsement and interest rate movements. (There is a lag of several months, typically, between loan closing and insurance endorsement.) As mortgage interest rates rise, claim rates on mortgages originated in that particular time period rise similarly. While not available in this data, this relationship exists when there are changes within individual years. For example, when interest rates dropped sharply in the first half of 1986, FHA's claim rate on loans originated in the last half of 1986 was much lower than the claim rate on loans originated earlier that year. Also illustrative of this pattern is the tremendous peak in 1981, when FHA/VA interest rates were stretching to the 17 percent mark, and claim rates hit an all-time high of 8.1 percent for 1981 originations.

Policy-year peaks

Chart 3 shows the number of claims that occurred by policy year for mortgages originated from 1981 to 1986. The increase in the number of claims for loans originated since 1983 stands out. This increase is tied to the growth in FHA activity that began in 1983 and continued steadily throughout the next several years.

Claims generally peaked in the third through fifth policy years and then declined in later years. Interestingly however, the 1979 and 1980 books of business are the exceptions to this rule. Chart 4 shows that in both years, there was a "second peak" for claims in the ninth policy year for 1979 originations and in the eighth policy year for 1980 originations. It is significant that 1987 was the year that both of these "second peaks" occurred.

Chart 5 provides claims by calendar year for loans originated from 1981 through 1986. In addition to the second peak for 1979 and 1980 origination years, the claims for the 1983 to 1985 books of business peaked in 1987. 1987 is also clearly a significant claim year for loans originated prior to 1980, as these loans also had a "second peak" in 1987.

Accordingly, it appears that the conditions existing in 1987, namely the regional recession/depression in the Southwest, may have been the cause for the upturn in claims on these seasoned loans. These 1987 peaks appear, in fact, to be an anomaly, rather than evidence of a new trend whereby a second wave of risk would be expected to occur in later policy years. Based on the FHA data, it appears that the third through fifth policy years have the greatest credit risk.

The occurrence of these second peaks in 1987, for seasoned loans that were at least 7 years old and may be as much as 17 years old, also indicates that whatever problems existed in 1987 had little, if anything, to do with the mortgage origination process. For a borrower to make mortgage payments for seven years or more before going into default is indicative of structural problems in the local economy rather than problems with the mortgage lending process.

Geographic analysis

There has been considerable discussion in recent years about the impact of location on the occurrence of claims. There is little doubt that regional economic conditions contributed greatly to the second peaks of claim years for many FHA-insured loans in 1987. In this section, claim rates by state are analyzed for FHA's 1972-to-1988 books of business.

To begin this analysis, five states with the most serious claim problems for three different endorsement years were analyzed to provide a perspective on the impact of the location of claims on FHA's overall claim performance. 1977 had the lowest annual claim rate; 1981 had the highest annual claim rate and 1986 had the most claims paid.

Table 2 provides the claim rates of states with the most serious problems for that book of business in 1977, 1981 and 1986. Each year shown in the table is noteworthy for significant feature. The 1977 list is represented predominantly by "Rust Belt" states. The 1981 list demonstrates the impact of the 1981-1982 recession on the Northwest and the beginning of the problems in the Southwest. The 1986 list shows the effect of the decline of states with oil-based economies.
 1977 Book of Business
 High Claim Rates
State Claims Mortgages Rate
Pennsylvania 266 1,970 13.5
Illinois 301 2,300 13.1
Indiana 471 4,354 10.8
Ohio 1,024 9,840 10.4
Michigan 308 3,209 9.6
5 State Total 2,370 21,673 10.9
Rest of U.S. 7,148 220,860 3.2
U.S. Total 9,518 242,533 5.9
5 State As % of U.S. Total 25% 9%
 1981 Book of Business
 High Claim Rates
State Claims Mortgages Rate
Oregon 409 966 42.3
Oklahoma 549 1,586 34.6
Illinois 1,345 4,054 33.2
Texas 4,672 16,256 28.7
Washington 862 3,084 27.9
5 State Total 7,837 25,937 30.2
Rest of U.S. 18,300 98,121 18.6
U.S. Total 26,137 124,058 21.1
5 States As % of U.S. Total 30% 21%
 1986 Book of Business
 High Claim Rates
State Claims Mortgages Rate
Alaska 1,374 4,605 29.8
Texas 12,694 88,334 14.3
Oklahoma 3,761 18,059 20.8
Louisiana 1,893 15,077 12.6
Colorado 7,404 58,220 12.7
5 State Total 27,126 184,295 14.7
Rest of U.S. 28,899 763,672 3.7
U.S. Total 56,027 947,967 5.9
5 States As % of U.S. Total 48% 19%

What makes the 1986 data so devastating for FHA is the percentage of all of its claims that came from these five oil-patch states. Texas, Oklahoma, Louisiana, Colorado and Alaska comprised 19 percent of FHA's origination business and 48 percent of its claims during 1986. In the other three analyzed years, the percentage of all FHA claims that came from the five high-claim states was between 20 and 30 percent.

The impact of the problems in the oil-patch region is demonstrated in Chart 6. Throughout the entire period (1972 to 1988), the market share of the three states in the Southwest tied to the oil industry (Texas, Oklahoma and Louisiana) was between 12 and 18 percent of FHA's origination activity. Starting in 1978, there has been a steady increase in the percentage of FHA claims generated from these three states. The percentage peaked in 1986 when approximately one-third of all FHA claims on loans originated in 1986 came from these states.

The concentration is even more pronounced when the Houston area is analyzed separately. There are six field offices in Texas with Houston covering much of eastern Texas, and as much as 12 percent of all FHA claims on one year's book of business (1982) for the entire country came out of the Houston office.

The Denver office, which covers the entire state of Colorado, provides a similar picture for later years in the 1980s. As late as 1981, the Denver office had a lower claim rate than the national average. By 1986, however, just about one of every seven FHA claims that resulted from a loan originated during that year came from Colorado.

During the 1980s, FHA claims were increasingly concentrated in those states relying heavily on the oil industry. This data documents the impact of the economic devastation on the housing industry. While loan-to-value (LTV) data was unavailable, it has shown in the past that low LTV loans were not exempted from this problem. In fact, for 1985 and 1986 originations, LTVs below 80 percent in the five oil states had higher claim rates than LTVs above 95 percent in the rest of the country.

This data may also explain the second peaks that occurred on seasoned loans. While data on claims by policy years is not available on a per-state basis, the economic problems and resulting property depreciation that beset the oil-patch states may be a significant factor in the number of seasoned loans that went into foreclosure during the period from 1985 to 1987.

What does this data mean for

FHA's financial condition?

Last month, FHA released its fiscal-year 1990 financial statement, and it included positive signs that FHA's single-family program has turned the corner and is now in the midst of a complete recovery from the problems that plagued its operation in the 1980s. For the first time since 1987, the Mutual Mortgage Insurance Fund (MMIF), out of which the basic single-family program is operated, generated a net income of $66 million from its insurance operations. HUD acknowledged in the financial summary that this improvement was the result of a decline in single-family defaults and claims.

What is even more encouraging is that this improvement occurred even though an accounting adjustment made by Price Waterhouse in its 1989 audit, which slowed down the earning of income, remained in effect (See sidebar). As a result of that adjustment, even though FHA collected approximately $400 million more in premiums in fiscal year 1990 than it did in 1988 before the adjustment went into effect, it earned $325 million less in total income.

The impact of this change is seen in the growth of the unearned premium account, which as risen from $3.8 billion in fiscal year 1988 to $5.6 billion in fiscal year 1990. This $1.8 billion nest egg is, in effect, a deferred compensation plan for the FHA program.

Another indication of FHA's improving financial strength is that there was a $638 million increase in "net cash operations." In fiscal year 1989 there had been only a $44 million increase and in fiscal year 1988 there had been a $322 million decrease in cash operations. The major items of the cash-flow statement are total premiums collected plus proceeds from the disposition of HUD properties minus claim settlement payments, premiums earned and the decline in loss reserves.

It is this last item that gives yet another piece of good news from the fiscal year 1990 audit. Loan loss reserves were lowered $219.8 million in the fiscal year 1990 audit. This is the second straight year that FHA loss reserves were lowered. Because Price Waterhouse increased loan loss reserves by $1.2 billion in fiscal year 1988 there has now been a 58 percent ($694 million) decline from that figure.

The next chapter

It would have been nice to conclude this article by stating that FHA is on the road to complete recovery. As the saying goes, however, there is another chapter of this story yet to be written.

While HUD moved quickly to correct the abuses that occurred under Secretary Samuel Pierce with the enactment of the HUD Reform Act of 1989 and the implementation of numerous management improvements, there is now growing concern that the FHA reforms have gone too far. Specifically this pertains to the provisions of the 1990 housing bill that raise down payments and increase mortgage insurance premium costs for homebuyers. It is the latter point that raises questions about FHA's future solvency.

With the implementation of a one-half percent annual premium in addition to the 3.8 percent upfront MIP on loans closed July 1, 1991, HUD has increased the likelihood of adverse selection in the FHA program. Because of the increase cost, lower risk borrowers who were the backbone of the FHA program are now opting for private mortgage insurance because of the lower cost and less red tape. FHA is, therefore, in effect becoming the insurer of last resort.

While definitive conclusions can not be reached for several more months, the initial returns are alarming. Applications for FHA insurance have dropped 28 percent from April to August 1991.

Based on the improvements discussed in this article, it is ironic that it now appears that FHA faces greater risk to its future solvency from the implementation of measures developed in the spirit of reform than it does from the performance of FHA's existing portfolio.

The Price Waterhouse


In the fiscal year 1980 audit, the accounting firm of Price Waterhouse made an adjustment to FHA's financial statement to slow down the earning of premium revenue to cover possible increases in defaults. The company explained its position as follows:

This change results principally

from applying more conservative

assumptions with

respect to house price appreciation,

and thus provides a better

method of matching premium

revenues to losses from single-family


Instead of earning revenue in the early years (policy years two through five), Price Waterhouse believes that HUD needs to defer income to cover possible defaults in the future.

On March 8, 1989, Price Waterhouse had stated in a GAO report that HUD was not recognizing premium income fast enough. In one year, Price Waterhouse shifted its position 180 degrees.

PHOTO : CHART 1 FHA Claim Rates for First Three Policy Years

PHOTO : CHART 2 Interest Rate and FHA Claim Rate

PHOTO : CHART 3 FHA 203(b) Claims Paid by Policy Year (1981-86 Originations)

PHOTO : CHART 4 FHA 203(b) Claims Paid By Policy Year (1979-1980 Origination)

PHOTO : CHART 5 FHA 203(b) Claims By Calendar Year of Claim

PHOTO : CHART 6 Percentage of FHA U.S. Total Insurance Written and Claims by Origination Year for Texas, Oklahoma & Louisiana 1972-1988
COPYRIGHT 1991 Mortgage Bankers Association of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:includes related article; analysis of the Federal Housing Administration's insurance and claims data
Author:Chappelle, Brian
Publication:Mortgage Banking
Date:Oct 1, 1991
Previous Article:Lending a helping hand.
Next Article:The new RESPA police.

Related Articles
FHA's roots.
A white paper on FHA refinancings.
Credit risk and the provision of mortgages to lower-income and minority homebuyers.
Distribution of credit risk among providers of mortgages to lower-income and minority homebuyers.
HUD proposes higher FHA mortgage limits.
How do you rate with FHA?
Refinance candidates benefit from FHA mortgage financing.
More work for builders as Clinton pushes laws.
AAMG pays $41 million FHA settlement.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters